PDA

View Full Version : Help with mr02 PCB


locwan
2006.10.12, 01:38 AM
I've burnt my fets on one of my mr02 PCB, and promtly replaced the burnt fets with some stock fets which I had laying around. I know for certain that the fets are in working order, and I'm experienced with replacing fets.

However, when I was done with the replacment, I had no forward, only reverse. It doesn't look like this is due to bad fets, as I've tried different combination of fets.

To clarify, when I hit forward on the TX, I can hear and see the motor trying to move somewhat, as if the power isn't getting through. The reverse signal works just fine, regardless the motor wires' orientation.

I've also checked that the fet pads are all there in good condition.. just can't figure out what's wrong. Anyone experienced this? Help please.

hesstoys
2006.10.15, 08:20 PM
Hi,
Although I'm new to all this, I may be able to help since I'm an electronic engineer. I just got my 11 year-old son involved with this type of RC racing and decided to learn this area of electronics. I've worked with FETs for many years and I'm quite familiar with on-time/off-time and on-resistance etc (also the many types from Vishay). From what others have told me, the idea of stacking FETs is to lower the on resistance. This is good but you need to be careful on the type of FET package. Some have two N-Channel or two P-Channel or one N and one P-Channel. Also the orientation of the number one pin. If you give me the part number I can tell you the type and then work from there.

Vic

Spoon
2006.10.15, 08:44 PM
The stock FET's that people are referring to are most likely the 3004's. There is a slim chance that they are 3010's. The upgraded FET's are most like 4562's.

locwan
2006.10.15, 10:51 PM
Thanks for the offer to help. To clarify, the stock fets I'm refering to are 3004's, and have only 1 p-channel and 1 n-channel for each individual fet. I believe the orientation of the fets on the fet pads are correct, or else there would be absolutely no response at all for both forward and reverse signals. Again, I have full reverse working, but with forward, the motor does give a response, and is struggling to turn, but it seems the power isn't getting through enough for it to function properly. I can only suspect that a connection is servered somewhere on the pcb or is connecting poorly.. but I haven't a clue where.

Also, I've checked that pins 1 to 3 of each fets is connecting okay by using a multimeter to check for continuity to it's respective vias, but the only via that gives a response to pin 4 is right underneath each individual fet, so I'm not able to check those for a proper connection.

byebye
2006.10.16, 02:04 AM
Did you try a different motor?

-Byebye

locwan
2006.10.16, 09:59 AM
Yes. Differernt motors were tried on. Also differernt fets as well, with the same result. Right now, I've switched the motor leads from + to -, and reversed the signal on the TX so that the car will run forward, but it is quite frustrating without the reverse (forward) signal, as I have to go pick up the car everytime I get snagged on the track.

byebye
2006.10.16, 11:31 AM
I suppose it's possible that you pulled one of the pads up on the forward side of the board. It's really not rocket science when changing FETs. You just have to make sure all feet are fully seated and nothing is shorted.

-Byebye

locwan
2006.10.16, 11:42 AM
Right. Although I do wish it was as simple as that.. I've done quite a few fet replacements and have become quite good in terms of soldering quality. If the fets are shorting, then they would likely overheat and blow the instant I put in the batteries.

Hesstoys, is it possible to try to use jumper wires to connect the n channel to the p channel directly from one fet to the other, and vice versa?

hesstoys
2006.10.16, 10:10 PM
Since you clearly know that the FETs are OK, then the signal (voltage) to the GATE is either not available at all (I believe from the PIC processor - damaged from shorted GATE to Drain) or you have an open trace. Today, I replaced the Hatachi HAT3004 with Si4542. Unfortunately I did not have the "62s. The 4542 are much better than the original in many respects. By the way, I stacked two+two. Perhaps next week I will have the time to fingerprint the control lines with my four channel digital scope and give you some voltage readings. Right now, my son won't let me fiddle with it. If you have a Digital Volt-meter, check the GATE voltage ( think its pin 1 & 3). Go online get the PDF data sheet for the Vishay Si4542DY.

Hesstoys

hesstoys
2006.10.16, 11:00 PM
Sorry, I should have read all the other postings..
First I need to correct myself. The GATE pins are 2 & 4. Not having a schematic, I pretty much know how there little babies work to drive the motor. I can explain in detail how it should work if you want to know. Anyway, I truly believe the drive signal to either pin 2 or 4 (check both sides - chips) is very low, possibly only 1 volt. If you have the 4562, they start conducting around 2 volts. The 3004 don't start unit the voltage reaches 3 volts. Now this is not full conduction. The 3004 must have 4-volts to fully turn on. How did this mess happen in the first place? My professional opinon for what its worth goes like this: Due to high current/overheating or inductive kickback (counter EMF) from the inductive motor (possibly a reverse-blocking Diode opened). a high-voltage inductive spike traveled down the GATE pin since the device was shorted, to the Microchip PIC processor I saw on my son's PCB.

Well, point is, you will need to make measurements or purchase a new PCB. Now, if you know someone with a burned-out PCB, take out the PIC and place it in yours. There is another complicated way of testing the FET/Motor combo but it will require pull-up/pull-down resistors and appling your own 4-5 volts on the GATE of the N-CH and -4 / -5 volts to the P-CH.

Vic

hesstoys
2006.10.16, 11:21 PM
Boy I'm a big knocklehead! I forgot to answer your question. Remember, we are dealing with two chips (IC). To begin, on either chip, the SOURCE and DRAIN are connected - four pins soldered with RED wire to motor, other chip with BLACK wire to motor. To move FORWARD, say the RED needs to be positive, N-CH - which should have the SOURCE grounded, is turned off. The P-CH, with the DRAIN connected to 5 volts, needs to be turned on with a negative signal. Now the other FET pak must provide the ground. Here, things are reversed, the N-CH is turned on with a positive signal to the GATE which will provide the ground between DRAIN and SOURCE; the P-CH must be off. If you jump any leads other than the way it is designed, you release the "smoke" from the IC. Once the "smoke" is gone, all electronic parts no longer work!

Last statement is a very old joke.

Vic

byebye
2006.10.17, 01:46 AM
Hesstoys it really sounds like you know what you are talking about. Glad to have you on the forums.

@locwan-When I said it was simple I just meant in terms of what is being done and why. I think this is just the one that failed ya know? I've had this happen on a stack and replaced the FETs again with a single layer of 4562's and all was well again.

At least now you have some pro electronics help from hesstoys and a board to toy around with. Hope you get it up and running.

-Byebye

locwan
2006.10.17, 01:24 PM
Thanks all for the abundance of info. If the pic processor is in fact damaged, then I will have no choice other then to buy another board, as I neither have another fried board with a good pic nor the proper eletronic equipment to do proper measurements and testing.. :( But these are good info that might become handy in the future.

I agree with Byebye, it's good to have you on the forums Hesstoy.

hesstoys
2006.10.17, 09:11 PM
Hi All,
Thanks for the nice words. Even with my 33 years in the electronic field, I sc--wed up. Although all the FET info is fact, I changed the wrong set...HEE HEE! Since my son had removed the motor, I thought the steering motor was the driving motor. So, forget the comment about soldering the RED and BLACK wire. On our PCB, the wires are screwed on. Anyway, I did the correct FETs today and the car runs very fast. Now my son needs to control his speed going into the turns. We go to an indoor track thats 12ft x 24ft. Anyway, I took pictures of the PCB with and without the FETs and can show you which veers connect to which FET pins. If anyone needs this info, shot me an email at hesstoys@hvc.rr.com. Next, I'm going to take scope measurements for anyone that may need it. Also, when working on these PCBs, its best to remove all wires (make a drawing) because if many strands break, it would be hard to push all that current thru the remaining tiny few.

vic